On Mon, Jan 5, 2015 at 2:19 PM, Mark Kettenis <[email protected]> wrote:
>> From: sven falempin <[email protected]>
>> Date: Mon, 5 Jan 2015 13:29:41 -0500
>>
>> On Thu, Dec 11, 2014 at 7:47 AM, Stuart Henderson <[email protected]> wrote:
>> >
>> > On 2014/12/11 07:41, sven falempin wrote:
>> > > Thank you all,
>> > >
>> > > I grep(ed) -r NCT6 in sys and didn't find wbsio, I guess those
>> > > christmass holydays will be much welcome !
>> > >
>> > > Does the wbsio detect the watchdog in the apu card ?
>> >
>> > No, as the manual says, "Only the hardware monitoring function is
>> > currently supported", and at present, and this driver doesn't attach
>> > on the APU.
>>
>>
>>
>> First of all , Happy new gregorian Year 2015 !
>> I hope you all survived the longest night of the year :-)
>>
>> Jan 5 12:36:56 unicornD /bsd: HOIHOIHOI wbsio_probe: id 0xc4
>> Jan 5 12:36:56 unicornD /bsd: >>> probe for wbsio* clone into wbsio0
>> Jan 5 12:36:56 unicornD /bsd: wbsio0 at isa0 port 0x2e/2 rev 0x52
>> Jan 5 12:36:56 unicornD /bsd: >>> probing for lm*
>> Jan 5 12:36:56 unicornD /bsd: >>> lm probe returned 0
>>
>> Nothing really interesting so far , but I can read the revision number .
>>
>> What's next ?
>
> Hmm, chris@ just added the NCT5104D chip, which happens to have the
> same ID :(. That will conflict with this diff.
To get thing straight the driver should read the complete ID, it 16 bit length :
wbsio0 at isa0 port 0x2e/2: NCT6106D: complete devid? 0xc452 rev 0x52
Index: wbsio.c
===================================================================
RCS file: /cvs/src/sys/dev/isa/wbsio.c,v
retrieving revision 1.8
diff -u -p -r1.8 wbsio.c
--- wbsio.c 1 Jul 2012 02:15:09 -0000 1.8
+++ wbsio.c 7 Jan 2015 16:01:09 -0000
@@ -51,6 +51,7 @@ struct cfattach wbsio_ca = {
sizeof(struct wbsio_softc),
wbsio_probe,
wbsio_attach
+ //?? ,wbsio_activate
};
struct cfdriver wbsio_cd = {
@@ -112,6 +113,7 @@ wbsio_probe(struct device *parent, void
case WBSIO_ID_W83637HF:
case WBSIO_ID_W83697HF:
case WBSIO_ID_NCT6776F:
+ case WBSIO_ID_NCT6106D:
ia->ipa_nio = 1;
ia->ipa_io[0].length = WBSIO_IOSIZE;
ia->ipa_nmem = 0;
@@ -130,7 +132,7 @@ wbsio_attach(struct device *parent, stru
struct isa_attach_args *ia = aux;
struct isa_attach_args nia;
u_int8_t devid, reg, reg0, reg1;
- u_int16_t iobase;
+ u_int16_t edevid,iobase;
/* Map ISA I/O space */
sc->sc_iot = ia->ia_iot;
@@ -173,8 +175,19 @@ wbsio_attach(struct device *parent, stru
case WBSIO_ID_NCT6776F:
printf(": NCT6776F");
break;
+ case WBSIO_ID_NCT6106D:
+ printf(": NCT6106D");
+ break;
+ default:
+ printf(": devid? 0x%02x", devid);
+ break;
}
+ /* Read complete device ID */
+ edevid = wbsio_conf_read(sc->sc_iot, sc->sc_ioh, WBSIO_ID) << 8
+ | wbsio_conf_read(sc->sc_iot, sc->sc_ioh, WBSIO_ID + 1) ;
+ printf(": complete devid? 0x%04x", edevid);
+
/* Read device revision */
reg = wbsio_conf_read(sc->sc_iot, sc->sc_ioh, WBSIO_REV);
printf(" rev 0x%02x", reg);
Index: wbsioreg.h
===================================================================
RCS file: /cvs/src/sys/dev/isa/wbsioreg.h,v
retrieving revision 1.3
diff -u -p -r1.3 wbsioreg.h
--- wbsioreg.h 1 Jul 2012 02:15:09 -0000 1.3
+++ wbsioreg.h 7 Jan 2015 16:01:09 -0000
@@ -43,6 +43,7 @@
#define WBSIO_ID_W83637HF 0x70
#define WBSIO_ID_W83697HF 0x60
#define WBSIO_ID_NCT6776F 0xc3
+#define WBSIO_ID_NCT6106D 0xc4
Shall i go from current and remove the 8 bit id ? shall we read the
second byte only if the first is a NCT device ?
>
> Seems the NCT5104D is basically a stripped down NCT6106D that has just
> the watchdog timer, 4 UARTs and GPIO. Perhaps some sort of probe is
> needed to see if a logical device is actually there. I guess we
> already have one in place for the hardware monitoring function since
> we check the I/O address before attaching lm(4).
>
--
---------------------------------------------------------------------------------------------------------------------
() ascii ribbon campaign - against html e-mail
/\
